Tool use to crow about

Crows have an unspoken motto: Sticks and stones may break my bones, but they sure can come in handy. Two new studies unveil the ability of at least some crows to use tools in sophisticated ways, without training, to obtain food.

Crows’ problem-solving feats in these studies underscore a substantial intelligence that has attracted relatively little scientific attention, according to both research teams.

Scientists have previously noted tool use among members of the crow family, or corvids, including dropping stones on intruders or prey and using paper as a rake and sponge. But few birds display comparable behavior, and researchers have largely concentrated on the extensive, flexible tool practices of chimpanzees and other nonhuman primates. Evidence from apes and monkeys, as well as other large-brained, social animals such as dolphins, has debunked the traditional view that tool use is a defining human characteristic.

In one study, appearing online August 6 and in the Aug. 25 Current Biology, zoologists Christopher Bird of the University of Cambridge in England, and Nathan Emery of Queen Mary, University of London report that captive rooks — members of the crow family — rapidly learn to use stones to obtain food on their own. The rooks dropped stones into a tube containing water in order to raise the water level and bring a floating worm within reach. This behavior recalls one of Aesop’s fables, in which a thirsty crow plunked rocks into a pitcher to raise the water level high enough for a drink.

All four rooks tested used stones provided by the researchers to raise the water level. Two birds succeeded on their first try. The other two needed two tries.

In all successful cases, birds put in only the number of stones needed to snatch the worm. Most of the time rooks did not begin by dropping the biggest stones into the water, but the birds quickly learned to do so.

The four rooks also quickly learned not to drop stones into a tube containing sawdust with a worm on top.

Although no evidence indicates that rooks use tools in the wild, the animals possess general mental abilities that they can apply to tool use, Bird and Emery propose. These include insights about physical rules, such as water in a container rising after a stone gets dropped in. In the team’s view, such insight fostered the birds’ success in the new study.

“Rooks do not use tools in the wild because they do not need to, not because they can’t,” Bird says. They get everything they need in their natural habitat without having to resort to devising tools.

In a comment slated to be published with Bird and Emery’s study, zoologists Alex Taylor and Russell Gray, both of the University of Auckland in New Zealand, note that the birds don’t have to have a plan in mind when they begin the task. The rooks in the study had already learned on their own to drop stones into a tube in order to collapse a platform inside and release food. Based on that experience, crows could have decided right away to try dropping a stone into the water-filled tube. After seeing the water rise and the worm move closer, birds would have added more stones.

Other researchers reported in 2007 that, in similar fashion, orangutans collected water from a drinking container in their mouths and spat it into a tube to raise the water level so that they could snatch a peanut floating on the surface.

It’s unclear whether orangutans and rooks devise a mental plan before acting or simply have the capacity to notice when a random act causes food to move closer and then repeat the act, comments experimental psychologist Amanda Seed of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany.

In a paper published online August 4 in PLoS ONE, zoologist Joanna Wimpenny of the University of Oxford in England and her colleagues report that captive New Caledonian crows can manipulate three wooden sticks, one at a time and in the correct sequence, to obtain food. Three-step tool use didn’t require, but was enhanced by, earlier training with the sticks.

New Caledonian crows have been shown to modify twigs and leaves to remove insects from crevices in the wild (SN: 3/22/03, p. 182). In earlier work directed by Oxford’s Alex Kacelnik, a New Caledonian crow bent or unbent strips of metal as needed to reach pieces of food.

Five of seven crows in the new study managed on their own to grab an available, short stick with their beaks, use it to drag a longer stick from a transparent tube and then pull food out from another tube with the longer stick. Four crows succeeded on their first trial.

Further testing showed that four crows, including one that had no prior experience with the experimental set-up, frequently used three sticks in correct sequence to obtain food. Birds used a short stick to drag a longer stick out of a tube, and used the longer stick to drag an even longer one from another tube. The birds could reach the food only with the longest stick.

Each crow committed its own pattern of errors in performing the three-tool task. Crows perceived food depth and tool length with varying degrees of accuracy, leading to a range of errors, the researchers hypothesize.

“Crows made errors that would not be expected if they had full knowledge of the requirements of the task and an ability to plan a complete sequence of novel actions, but critically, it was not possible to say which ability was lacking,” remarks Seed.

Whatever the case, New Caledonian crows’ brains can generate the mental firepower needed for tool use, Gray says.

He and his colleagues have shown that New Caledonian crows have larger brains for their body size than most bird species, including other corvids. In an unpublished study, the researchers find that regions of the brain that integrate information have expanded in the New Caledonian crows.

Brain evolution, and thus the emergence of a tool-using capacity, probably proceeded independently in corvids and nonhuman primates, Gray notes.